trialswhose

trialswhose is a fictional term used in theoretical discussions of trial design and meta-research. It denotes a class of studies in which the decision to include data, define endpoints, or report results is conditioned on pre-specified relational criteria embedded in the protocol. The concept is intended as a thought experiment about how conditional reporting and outcome selection can influence conclusions drawn from experimental data.

Origin and status: The term has appeared in academic blogs, workshop notes, and speculative discussions about evidence generation. It is not part of formal nomenclature in clinical trial registries or peer-reviewed guidelines, and it is not tied to a specific disease area or regulatory framework.

Core features: Trialswhose are characterized by conditional data handling and reporting that depend on relationships defined in the protocol. They emphasize the potential impact of selective reporting, the interpretation of hierarchy among endpoints, and the robustness of conclusions to alternative analysis plans. The term is generally used to illustrate potential biases rather than to describe an accepted design practice.

Implications: As a concept, trialswhose highlights the importance of full pre-registration, pre-defined primary outcomes, and transparent reporting of all planned analyses. It also serves as a cautionary example against flexible or post hoc inclusion criteria that could bias estimates. Critics warn that the term may lack precise definitions and could obscure more rigorous concepts such as adaptive designs or informative censoring.

See also: publication bias, selective reporting, preregistration, open science, trial registries, outcome reporting bias.